Many diseases and/or other pathological processes or conditions are caused by dysfunction and/or disregulation of certain proteins. These disease-related proteins may have their structures altered, relative to their “normal” or “wild-type” counterparts, and/or may be expressed in larger (up-regulated expression) or smaller (down-regulated expression) quantities in a given disease state, relative to “normal” physiological conditions. In some cases, proteins having altered structure and/or function may be used as markers associated with a particular human or animal disease, for instance, as a diagnostic for earlier detection of the disease, or the like. In many cases, the particular protein(s) of relevance to a given pathological process of a disease or other condition are unknown. Identification of such protein(s) would be useful for development of new diagnostic tests, or the like.
A general approach to the identification and characterization of protein markers is based on the analysis of protein compositions of samples of biological material (biological fluids, such as blood, serum, plasma, cerebrospinal fluid, tissues, cells, etc.) using high resolution separation techniques. For instance, proteins isolated from control and experimental samples or populations can be subjected to proteolytic cleavage, and their cleavage products identified using liquid chromatography (LC) coupled with tandem mass spectrometry (LC-MS-MS). Many protein separation techniques are based on multi-dimensional separation of proteins from a sample, typically by two-dimensional gel electrophoresis (2-DE) or two-dimensional high-performance liquid chromatography (2D-HPLC). The two-dimensional (2-D) protein maps for pathological samples may be obtained and compared with those for reference samples; positions of proteins observed as “spots” on (2-DE) maps or as “peaks” on 2D-HPLC maps can be compared, and those that are present (or absent) in the maps obtained from pathological samples but absent (or present) in the maps obtained from the reference samples may be judged as being likely to correspond to pathologically relevant proteins. Additionally, quantities of proteins estimated as intensities of the spots (or peaks) may be evaluated and compared between the pathological and reference samples. Those that are significantly different may be considered as pathologically relevant in some cases.
It has also been recently established that a pattern of the presence/absence and/or the relative quantities of multiple proteins (a “signature”) may also be of diagnostic relevance, where the proteins judged to be of interest are identified by peptide mapping and/or mass spectrometry. Mathematical or statistical techniques, such as pattern recognition techniques, can be used to analyze the pattern produced by these experimental techniques and produce a diagnostic classification. However, this approach is often highly inefficient, for example, due to the inherent necessity of analyzing all of the proteins in a given sample, whereas only a small portion of the proteins may have any pathological relevance.
Several different methods for reducing the analytical complexity of protein mixtures have been developed. These methods are typically based on fractionation of the original mixture prior to 2D analysis by gel electrophoresis or 2D-HPLC. One such method is the separation of proteins by the technique of free-flow electrophoresis. However, this technique, while fractionating the original protein mixture, may result in multiple 2D analysis of simplified fractions, i.e. while reducing the complexity of analysis and improving resolution, it inherently greatly increases the number of samples where further analysis is required.
Another method is fractionation based on the affinity of proteins to different natural ligands and/or pharmacological compounds; however, this approach, while allowing separation of proteins according to protein functions, may result in a large increase in the number of samples for further analysis, and often requires additional knowledge or presumption concerning the differences between the samples.
One disadvantage of most fractionation techniques is that they generally cannot preserve protein-protein or protein-ligand interactions. Differences among biological interactions are often important for elucidating and detecting changes among samples. Additionally, most of the fractionation techniques rely on separation due to a fixed physical attribute, such as molecular size or net charge. While these attributes may be very important for distinguishing among biomolecules in a complex mixture, they generally do not cover all of the potential differences between biomolecules representing, e.g., normal vs. disease states, differences in configuration etc. Another important disadvantage of present fractionation techniques is related to their inability to separate mixtures based on differences between structural changes in, e.g., glycosylation patterns and/or conformational changes. These changes are often important for identifying proteins that either participate in and/or are the result of a disease state. For example, if a protein is misfolded as a result of genetic mutation, the net charge and size of the protein may not vary significantly, and more importantly, the protein's expression level might be the same for the underlying normal vs. disease states. Finally, natural genetic variability among individuals can significantly contribute to a very large scatter in the expression levels (concentrations) of biomolecules in a biological sample. This variability generally necessitates use of statistically large number of samples to robustly detect differences innate to a particular pathological condition, rather than to genetic variability. Natural genetic variability is often a significant hindrance in implementing protein marker based diagnostics due to reduction of the sensitivity and/or specificity of the diagnostic test.
While significant advances in the field of molecular and/or sample characterization have been made, improvements are therefore needed to add specificity, versatility, convenience, and/or improve efficiency.